google/gemma-3-4b-it
Image-Text-to-Text
•
Updated
•
624k
•
441
Gemma 3 Release
-
-
google/gemma-3-4b-pt
Image-Text-to-Text • Updated • 58.8k • 66 -
google/gemma-3-1b-pt
Text Generation • Updated • 145k • 106 -
google/gemma-3-1b-it
Text Generation • Updated • 1.63M • 320 -
google/gemma-3-12b-pt
Image-Text-to-Text • Updated • 49.3k • 44 -
google/gemma-3-12b-it
Image-Text-to-Text • Updated • 348k • 321 -
google/gemma-3-27b-pt
Image-Text-to-Text • Updated • 20.5k • 90
google/gemma-3-27b-it
Image-Text-to-Text • Updated • 783k • • 1.2kNote ^ transformers-based pre-trained and instruct models
google/shieldgemma-2-4b-it
Image-Text-to-Text • Updated • 3.14k • 82Note ^ ShieldGemma 2
-
google/gemma-3-4b-it-qat-q4_0-gguf
Image-Text-to-Text • Updated • 9.46k • 91 -
google/gemma-3-4b-pt-qat-q4_0-gguf
Image-Text-to-Text • Updated • 412 • 14 -
google/gemma-3-1b-it-qat-q4_0-gguf
Text Generation • Updated • 3.28k • 27 -
google/gemma-3-1b-pt-qat-q4_0-gguf
Text Generation • Updated • 160 • 6 -
google/gemma-3-12b-it-qat-q4_0-gguf
Image-Text-to-Text • Updated • 6.75k • 82 -
google/gemma-3-12b-pt-qat-q4_0-gguf
Image-Text-to-Text • Updated • 398 • 12 -
google/gemma-3-27b-it-qat-q4_0-gguf
Image-Text-to-Text • Updated • 63.6k • 181
google/gemma-3-27b-pt-qat-q4_0-gguf
Image-Text-to-Text • Updated • 394 • 18Note ^ GGUFs to be used in llama.cpp and Ollama. We strongly recommend using the IT models.
-
google/gemma-3-4b-it-qat-q4_0-unquantized
Image-Text-to-Text • Updated • 12 • 3 -
google/gemma-3-27b-it-qat-q4_0-unquantized
Image-Text-to-Text • Updated • 4 • 5 -
google/gemma-3-12b-it-qat-q4_0-unquantized
Image-Text-to-Text • Updated • 6 • 5 -
google/gemma-3-1b-it-qat-q4_0-unquantized
Text Generation • Updated • 9 • 1 -
google/gemma-3-4b-it-qat-int4-unquantized
Image-Text-to-Text • Updated • 2 • 1 -
google/gemma-3-12b-it-qat-int4-unquantized
Image-Text-to-Text • Updated • 2 • 4
google/gemma-3-1b-it-qat-int4-unquantized
Text Generation • Updated • 2Note ^ unquantized QAT-based checkpoints that allow quantizing while retaining similar quality to half precision